Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 23469
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal appeals and court procedures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ActionTypes
    -0.15
    .builders
    -0.15
     actionTypes
    -0.14
    awei
    -0.13
    eldo
    -0.13
    iversit
    -0.13
    ERGY
    -0.13
    анÑĸÑı
    -0.13
    GENCY
    -0.13
    jab
    -0.13
    POSITIVE LOGITS
     racing
    0.41
     race
    0.39
     Racing
    0.38
     rac
    0.37
     horse
    0.36
     horses
    0.35
     races
    0.35
     raced
    0.34
    horse
    0.32
     Race
    0.32
    Activations Density 0.016%

    No Known Activations